Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CardioSAM: Topology-Aware Decoder Design for High-Precision Cardiac MRI Segmentation

Ujjwal Jain

Abstract

Accurate segmentation of cardiac structures in cardiovascular magnetic resonance (CMR) images is essential for reliable diagnosis and treatment of cardiovascular diseases. However, manual segmentation remains time-consuming and suffers from significant inter-observer variability. Recent advances in deep learning, particularly foundation models such as the Segment Anything Model (SAM), demonstrate strong generalization but often lack the boundary precision required for clinical applications. To address this limitation, we propose CardioSAM, a hybrid architecture that combines the generalized feature extraction capability of a frozen SAM encoder with a lightweight, trainable cardiac-specific decoder. The proposed decoder introduces two key innovations: a Cardiac-Specific Attention module that incorporates anatomical topological priors, and a Boundary Refinement Module designed to improve tissue interface delineation. Experimental evaluation on the ACDC benchmark demonstrates that CardioSAM achieves a Dice coefficient of 93.39%, IoU of 87.61%, pixel accuracy of 99.20%, and HD95 of 4.2 mm. The proposed method surpasses strong baselines such as nnU-Net by +3.89% Dice and exceeds reported inter-expert agreement levels (91.2%), indicating its potential for reliable and clinically applicable cardiac segmentation.

CardioSAM: Topology-Aware Decoder Design for High-Precision Cardiac MRI Segmentation

Abstract

Accurate segmentation of cardiac structures in cardiovascular magnetic resonance (CMR) images is essential for reliable diagnosis and treatment of cardiovascular diseases. However, manual segmentation remains time-consuming and suffers from significant inter-observer variability. Recent advances in deep learning, particularly foundation models such as the Segment Anything Model (SAM), demonstrate strong generalization but often lack the boundary precision required for clinical applications. To address this limitation, we propose CardioSAM, a hybrid architecture that combines the generalized feature extraction capability of a frozen SAM encoder with a lightweight, trainable cardiac-specific decoder. The proposed decoder introduces two key innovations: a Cardiac-Specific Attention module that incorporates anatomical topological priors, and a Boundary Refinement Module designed to improve tissue interface delineation. Experimental evaluation on the ACDC benchmark demonstrates that CardioSAM achieves a Dice coefficient of 93.39%, IoU of 87.61%, pixel accuracy of 99.20%, and HD95 of 4.2 mm. The proposed method surpasses strong baselines such as nnU-Net by +3.89% Dice and exceeds reported inter-expert agreement levels (91.2%), indicating its potential for reliable and clinically applicable cardiac segmentation.

Paper Structure

This paper contains 20 sections, 15 equations, 6 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Proposed Methodology
  3. Formal Problem Definition and Topological Overview
  4. Foundation Model Backbone: Feature Manifold Extraction
  5. Cardiac-Specific Attention Module (CSAM)
  6. Advanced Contextual Gating Modules
  7. Multi-Scale Decoder with Spatial/Channel Attention
  8. Gradient-Based Boundary Refinement Module (BRM)
  9. Composite Objective Function Optimization
  10. Experimental Details
  11. Dataset Details and Pre processing
  12. Baselines
  13. Metrics
  14. Implementation Details
  15. Results and Discussions
  16. ...and 5 more sections

Figures (6)

  • Figure 1: Class Distributions
  • Figure 2: CardioSAM architecture diagram with all modules and data flow
  • Figure 3: Cardiac MRI medical images (ACDC dataset). (a) Normal Cardiacs; (b) Previous Myocardial Infarction; (c) Hypertrophic Cardiomyopathy; (d) Dilated Cardiomyopathy; (e) Abnormal Right Ventricle.
  • Figure 4: Learning Rate Sensitivity
  • Figure 5: CardioSAM Segmentation Results on ACDC Test Set
  • ...and 1 more figures